A simple corn with labor input-output model [e.g., von Weizsacker 1971] is used in this chapter to present and analyze the modern treatment of the Marxian labor theory of value and exploitation. There is one produced good, corn, and homogeneous labor. The input-output technology for each enterprise is specified by:

A = number of bushels of seed corn per bushel of harvest corn, and

a = number of hours of labor needed per bushel harvest corn.

A firm's gross output of X bushels of corn requires AX bushels of seed corn as an input and L = aX hours of labor. For simplicity (not realism), it is assumed that no other scarce inputs are required. For the technology to be viable, it is assumed that A < 1, i.e., less than a bushel is needed to grow a bushel. It takes one time period, called a "year," for the labor L = aX to produce the output X by using up the inputs AX. The inputs are required at the beginning of the year and the outputs are available at the end of the year.

Let p*= money price of corn per bushel, let w = money wage rate (at year-end), and let r = rate of interest for the year. Wages are paid at the end of the year while corn inputs are purchased at the beginning of the year. The capital outlay per unit output is p*A. If that capital is deposited in a savings account, it will compound to (1+r)p*A at the end of the year. That is the passive use of capital. Alternatively, if the capital is used "actively" by being invested in production, then one unit of output will be produced (no uncertainty), sold for p*, and then Labor is paid the wage wa. In a perfectly competitive model with no uncertainty, capital can be switched freely between the passive and active uses, so competitive arbitrage will enforce equality in equilibrium between the passive return of (1+r)p*A and the active return of p* - wa:

Competitive Equilibrium Condition: p* = wa + (1+r)p*A.

Solving for the equilibrium price yields:

p* = wa[1–(1+r)A]-1.

Dividing through by the money wage rate w expresses the price p = p*/w in terms of the numeraire of labor:

p = p*/w = a[1–(1+r)A]-1.

After more than a century of analysis and interpretation, Marxian economics has arrived at a precise modern formulation of a Marxian labor theory of value and exploitation [e.g., Morishima and Seton, 1961; Okishio, 1963; Morishima 1973; Wolfstetter, 1973]. It is this modern theory that is analyzed here. The theory stands by itself. We are not concerned here with the question of whether or not this "Marxian labor theory of value and exploitation" represents "what Marx really meant."

There are several ways to state the definition of the Marxian labor value v of a unit of corn. From the neo-classical viewpoint, it is the equilibrium price of a unit of corn if the interest rate ("rate of profit") is zero. The equilibrium price (in terms of labor) was

p(r) = a[1–(1+r)A]-1

so we have

Marxian Labor Value = v = p(0) = a[1–A]-1.

There is also the "net product" definition. Given the gross product X, the required seed corn was AX so the net product is defined as Y = X – AX (note how the definition subtracts seed corn from harvest corn "as if" there was no time difference). Then the Marxian labor value v could be defined as the labor necessary to produce one unit of net output. If net output = Y = 1 = X – AX, then the required gross product is

X = [1–A]-1

so the required labor is

v = a[1–A]-1.

The "historical" definition is based on the summation of all the labor directly and indirectly embodied in a unit of corn (assuming constant technology throughout the past). One unit of corn requires the direct labor of a units. It required the seed corn A which required the labor aA. The seed corn A also required the seed corn AA, which required the labor aAA, and so forth. Summing the labor

v = a + aA + aA2 + aA3 + ... = a[1 + A + A2 + A3 + ...]

= a[1–A]-1

(using the formula for the geometric series to evaluate the sum for 0<A<1). Note that this definition adds together labor from different time periods "as if" there was no time difference. All the definitions of labor value are equivalent [see Wolfstetter 1973 for yet another equivalent definition].

Each unit of labor is paid the money wage w so the physical wage in terms of corn is the

Wage-Basket = z = w/p* = 1/p.

One could think of labor as being paid z bushels per unit of labor. When a worker expends one unit labor, the payment is the wage-basket z with the Marxian value (or "labor content") vz. This is called the

Necessary Labor = Paid Labor = vz.

The remainder is called the

Surplus Labor = Unpaid Labor = 1 – vz.

The ratio of surplus labor over necessary labor is the

Marxian Rate of Exploitation = e = (1 – vz)/vz.

Morishima [1973, see Ellerman 1983 for a proof in the simple one commodity model] proves the:

Hence the Marxian theory concludes that Labor is exploited if the rate of interest or rate of profit is positive.

There are a few frequent misinterpretations of Marxian exploitation theory which must be mentioned first. The Marxian theory is not a bargaining power theory; the setting is a perfectly competitive model. Marx wrote extensively about the inequalities of the marketplace, but he wanted to criticize capitalist production itself, not just monopolistic imperfections. Hence he set out to expose exploitation in competitive capitalism, and the modern formulation preserves that competitive setting.

Marx also tried to relate the exploitation analysis to the workplace power relationship of the employer over the workers. In spite of the rhetoric which usually accompanies presentations of the modern theory, the role of power relations did not survive in the modern reformulation. No assumptions about power relations were made in the input-output model, yet the presence of Marxian exploitation can still be derived. Hence the modern result does not depend on power relationships. The veneer of rhetoric about "the capitalist forcing the workers to work longer than it takes to produce their labor-power" (wage-basket z) only obscures the real basis for the exploitation result.

The definition of Marxian value v systematically neglects the effect of time–an effect registered by the interest rate. Time puts a difference on commodities. As any farmer could testify, having corn available to plant at planting time is quite different from having the otherwise identical corn at harvest time. The seed corn and harvest corn are economically distinct–like "apples and oranges." One unit of a commodity at time t is equivalent to 1+r units at time t+1 in the sense that the market will trade one for the other (at constant prices). For example, the loan market trades $1 for $(1+r) a year from now. The Marxian value definition treats the units of labor (or corn) at different times as being the same (so they can be meaningfully added together), and thus the definition implicitly treats the interest rate as being zero.

Consider the net product definition of Marxian value v. The definition of the net product y = X – AX assumes that the beginning-of-the-year inputs AX are commensurate with the end-of-the-year outputs X so that the former can be subtracted from the latter to arrive at the net product. However, the difference X – AX is as meaningful as the difference "4 apples minus 3 oranges." The inputs AX are equivalent to (1+r)AX units at the end of the year, so the time-corrected net product in terms of commodities timed with outputs is:

y(r) = X – (1+r)AX.

When the corrected net product y(r) is equal to 1, the gross output is:

X = [1 – (1+r)A]-1.

Then the time-corrected Marxian value of a bushel of corn is:

v(r) = a[1 – (1+r)A]-1

in terms of beginning-of-the-year labor–which is precisely the price p(r) of a bushel of corn in terms of labor.

It is also possible to apply the time correction to the "historical" definition of Marxian value since that definition adds to labor performed in different time periods. If one bushel of corn is produced at year's end, then all the past embodied labor can be transformed into the equivalent beginning-of-the-year labor before being summed. The labor aAn performed n years before the beginning of the current year is equivalent to a(1+r)nAn units of labor at the beginning of the year. Hence the corrected historical definition of Marxian value is:

[assuming that (1+r)A<1] which is the same as the corrected net product definition.

What is the difference between the Marxian value v and the competitive market price p(r)? The difference is that Marxian value definition ignores time. Time is registered by the interest rate in the model. The uncorrected Marxian value v is p(0) the price when the interest rate is zero, and the corrected Marxian value v(r) is identical with the price p(r).

What happens to "exploitation" under the time correction? The time-corrected necessary labor contained in the wage-basket z paid for one unit of labor is:

v(r)z = p(r)z = 1

so the surplus labor is 1 – v(r)z = 1 – 1 = 0. Hence the exploitation result vanishes under the time correction.

The Fundamental Marxian Theorem ( e > 0 if and only if r > 0) is often interpreted as showing the exploitation is the hidden inner meaning of the charging of interest. Our results indicate that precisely the opposite is the case; the charging of interest is the hidden inner meaning of "exploitation."

The modern formulation of the Marxian labor theory of value and exploitation is in fact a just-price theory. It takes as a normative benchmark the time-saturated regime where the rate of interest (called the "rate of profit") is zero. It evaluates the transactions of the actual economic regime (where r is positive) at the benchmark prices. It finds that the workers receive less in the actual regime than they would in the benchmark regime; that difference is precisely the "exploitation." Of course, Marx did not intend or desire the theory to be only a just-price theory. But that is one of the ways a theory might fail. When finally worked out in a detailed and consistent form, the theory might fall far short of the original expectations.

Let r be the positive interest rate in the actual regime, so p(r) is the actual price of a bushel of corn in terms of the numeraire of labor. Since the price of labor in terms of labor is always unity, p(r)z = 1 so the real wage basket z = z(r) = 1/p(r) is also a function of r.

In the benchmark regime where r = 0, the just-price of corn is v = p(0) the Marxian value of a unit of corn. The just wage, which "represents the real wage rate that would prevail if there was no exploitation" [Morishima 1973, p. 54], is:

Just Wage = z* = z(0) = 1/p(0) = 1/v = [1–A]/a = Net Product per unit Labor.

As the interest rate r moves from zero to a positive value, the price of labor, the real wage z(r), decreases so labor-sellers are worse off in the actual regime in comparison with the benchmark regime. How much worse off? In selling the labor L = aX, the workers would receive z*L in the benchmark regime and they receive zL in the actual regime. The difference is:

[z(0)–z(r)]L = z*L – zL = ([1–A]/a – z)aX = X–AX – zL

= Net Product - Wage Corn

= Surplus Product.

Thus the so-called "surplus product" is just the difference between the "just corn wages" and the actual corn wages for the labor L. And the benchmark value of that wage differential in terms of labor is:

p(0)[z(0) – z(r)]L = (1–vz)L = v(X–AX) – vzL = Total Surplus Labor.

In the (hypothetical) transition from the benchmark to the actual regime, the economic position of labor-sellers worsened, and that is precisely the "Marxian exploitation." The difference in the wage-bill is the "surplus product" in terms of corn and the difference is the "surplus labor" in terms of labor.

The same sort of "exploitation" analysis could be applied to any price change. Here is an apple-selling example. Suppose in the benchmark situation,

Benchmark Prices: 10 Apples = 1 Bushel of Corn.

But in the actual situation, the price of apples dropped relative to corn.

Actual Prices: 15 Apples = 1 Bushel of Corn.

Suppose the apple-owner sells 300 apples in return for 300/15 = 20 bushels of corn. Let us "pierce the veil" of this competitive market transaction to "reveal its inner nature." In return for the 20 bushels, the apple seller first gives up 200 apples. The 200 apples have the same "value" as the 20 bushels (i.e., "value" = benchmark prices). Everything seems fair and square. The 200 apples were "paid for" by the 20 bushels. But then the apple-seller is "forced to alienate" an additional 100 apples which is "appropriated as a surplus" by the corn-owner without any further corn payment in return. These extra 100 apples are the "unpaid" apples. In terms of corn, the corn-owner gave up 20 bushels to receive the "value" of 30 bushels so the surplus appropriated by the corn-owner represented 10 bushels of corn.

The ratio of the unpaid apples to the paid apples is 100/200 = .5 so there is a 50% rate of exploitation. "Beneath the facade" of the market transaction, we have revealed the "exploitation" of the apple-seller by the "forced alienation" of the surplus apples.

Marxian exploitation theory applies this same methodology to the labor contract. It clearly has nothing to do with workplace power relations. The time-saturated benchmark regime defines the "just prices." The just interest rate is zero, the just corn price per bushel is v, and the just wage is z* = 1/v. In the hypothetical transition from the zero-interest benchmark model to the actual model, time enters as a scarce resource commanding a positive price (the positive interest rate). The price p(r) of corn in terms of labor is a monotonic increasing function of r:

p(r) = a[1 – (1+r)A]-1 = a[1 + (1+r)A + (1+r)2A2 + ...]

so there is more labor for the same corn, i.e., less corn for the same labor, than in the benchmark model. Where, say, 200 units of labor may have previously traded for 20 bushels, it now takes 300 units of labor to buy 20 bushels of corn. The first 200 units of labor have the same "value" (= benchmark price) as 20 bushels of corn, so the last 100 units of labor represent "unpaid labor" (at benchmark prices). Thus there is "exploitation" if and only if the interest rate is positive. That "Fundamental Marxian Theorem" may "be considered as the heart and soul of Marxian philosophy..." [Morishima 1973, p. 6].

The modern Marxian labor theory of value and exploitation has nothing to do with workplace power relations, with wage labor, or even with capitalist property relations. It turned out to be a pre-liberal Aristotle-Aquinas "interest grumble" dressed up in Marxian garb.

A simple corn and labor input-output model was used above to present and analyze the modern Marxian labor theory of value and exploitation. The same underlying technology is now used to present and analyze marginal productivity theory. Then all three theories (LTV, MP theory, and LTP) are compared in the same setting.

Marginal productivity (MP) theory claims that under competitive conditions:

(1) each worker produces his or her marginal product,

(2) each worker receives the value of his or her marginal product, and

(3) thus each worker gets what he or she produces.

A similar result would hold for each other factor.

The first difficulty in representing this argument in the corn and labor model is that the technology does not allow substitution between capital (seed corn) and labor. If labor is increased with no extra capital, there is no change in output (isoquants are right-angles with this Leontief technology). Hence we must define the marginal net product of labor [the classic treatment of MP theory with fixed coefficients is Georgescu-Roegen 1935]. We add an extra unit of labor and simultaneously add as much extra capital as is needed by the labor. Then we charge the extra capital against the increase in product. The net increase in output will be the marginal (net) product of labor.

One extra unit of labor (with enough extra capital) will produce 1/a extra units of output. The extra capital (seed corn) needed by 1/a units of output is A/a units of capital. One must be careful not to directly compare beginning-of-the-year corn (seed corn) with year-end corn (harvest corn) since they are separated by a time period. The market will exchange one unit of corn now for (1+r) units of corn at the end of the year. Hence the extra A/a bushels of seed-corn is equivalent to (1+r)A/a bushels of harvest corn. Charging the extra seed-corn against the extra product yields the:

Marginal Net Product of Labor = MNPL = 1/a – (1+r)A/a

= [1–(1+r)A]/a.

Multiplying through by the money price of seed corn p* yields

Thus the market value of the marginal net product of labor is equal to the wage rate so MP theory implies that each worker "gets what he produces." The use of the interest rate is necessary in the marginal productivity conditions when the input and output differ by a time period [e.g., see the Wicksell conditions in Samuelson 1937, p. 495].

The labor theory of property is a property theory, not a price theory. It is perfectly compatible with marginal productivity (MP) theory as a price theory (or any other price theory). However, MP theory has a central ideological role that far overshadows its price-theoretic function. That broader role is to show that each factor would "get what it produces" under competitive capitalism. In that role, MP theory does conflict with the labor theory of property. The conflict is not at the normative level; there is little disagreement that Labor should get what Labor produces. The questions are the factual questions of what Labor produces and what Labor gets.

Conventional economics imposes a certain preconceived "picture" on the firm, the picture of each factor supplier getting a certain distributive share of the output. Any remainder goes to a "residual claimant." This distributive shares or "shared pie" picture is a complete misrepresentation of the structure of property rights in a firm. One party legally appropriates 100% of the outputs, the positive product. The other legal parties who supply inputs appropriate 0% of the outputs. How can this be? How can one party appropriate 100% of the product when there are other scarce factors? Because that same party also appropriates 100% of the negative product, i.e., bears all the liabilities for the used-up inputs. The so-called "residual claimant" in fact claims the whole product. Thus the actual structure of property rights in production is one of complete asymmetry, not the symmetry of the distributive shares picture.

In the example, the gross output of harvest corn is the sum of the wage-bill plus the seed-corn (expressed in terms of harvest corn):

X = zL + (1+r)AX.

It is tempting to picture Labor as getting one share of the product with the seed-corn owner getting the other share. Labor is pictured as appropriating "Labor's share of the product" (zL,0,0) and the seed-corn owner as appropriating "Capital's share of the product" ((1+r)AX,0,0). There is no additional residual left for the residual claimant.

The actual structure of property rights is totally different. Hired labor appropriates no share of the product. The residual claimant, far from getting "nothing," appropriates the whole product (X,–AX,–L). Instead of being a co-claimant of the product, Labor is the party to whom the whole product appropriator is liable for the liability –L. That liability is satisfied by the corn-wage payment (zL,0,0) in return for the labor (0,0,L).

The distributive shares picture is false as a description of property relations, not as a description of value relations. Whether Labor appropriates the "share of the product" (zL,0,0) or receives (zL,0,0) as a wage payment, Labor still ends up with the income zL. Indeed there may be many different sets of property relations which yield the same set of value relations. For example, reverse the hiring relation. Let Labor buy the seed-corn (on credit) rather than the seed-corn owner hire Labor. Then Labor would appropriate the whole product (X,–AX,–L) and would net the Labor product (X,–AX,–L) + (0,0,L) = (X,–AX,0). The liability –AX is satisfied with the payment of ((1+r)AX,0,0) so Labor would again end up with the same value X-(1+r)AX = zL.

The two firms, one capitalist with Capital hiring Labor and the other a labor-managed firm with Labor hiring capital, are diametrically opposite in the structure of property rights (and in the structure of management control rights).

Figure: Different property appropriations with the same income distribution

In this model, the opposite property structures yield the same value relations. Labor and Capital have the same net income under each structure. The difference is property theoretic, not price theoretic. MP theory is not a property theory. It does not address property theoretic questions. It could not show that Labor appropriates what it produces because, in fact, hired-Labor appropriates none of its product. It is only in a metaphorical sense that Labor "gets" a share of the product (e.g., in the wage payment).

It is only an animistic metaphor to picture each factor as "producing" its marginal product. Capital does not "produce" its marginal product. Capital does not "produce" at all. Capital is used by Labor to produce the output. When capital is increased, Labor produces extra output by using up the extra capital. The "marginal product of capital" is the ratio of Labor's extra positive product over its extra negative product.

Neo-classical marginal productivity theorists are fond of observing that the theory applies symmetrically to any factor, e.g., "You can switch the roles of labor and land" [Samuelson 1976, p. 543]. Why can't one do the same in the labor theory of property? Why not define the "corn product" (X,0,–L) as the product of the corn input AX? In this "corn theory of property," the corn "produces" (X,0,–L) = (X,-AX,-L) + (0,AX,0), but the corn-supplier is only paid for the corn-input (0,AX,0) while some other party receives the difference (X,0,–L) – (0,AX,0) = (X,–AX,–L) which is the whole product.

The difficulty with this "corn theory" lies in a concept noticably absent from the neo-classical vocabulary, the concept of responsibility. Persons have the capacity for responsible agency; things don't. The only services which can be responsible for producing the whole product are the services of human beings, not the services of capital, land, or other commodities. That is why there is a labor theory of property, not a corn, land, or capital theory of property.

But what about the owners of the corn, capital, or land? They have the capacity for responsible agency. If a land-owner, for example, worked as a manager in an enterprise, then he or she would qualify for that reason as part of the responsible party Labor, not by reason of the land ownership.

The analysis of MP theory given here must be differentiated from the common "criticisms" of MP theory. MP theory is often criticized on the grounds that the actual economy is uncompetitive, that marginal products are difficult to measure, or that it does not justify the original distribution of factor ownership. These common arguments are often taken as "refuting" MP theory. Yet they really don't touch the core assertion of the theory, the assertion that competitive capitalism would allocate to "each according to what he and the instruments he owns produces" [Friedman 1962, pp. 161-162]. It is this core assertion which is incorrect.

The ideological importance of MP theory lies in the attempt to show that each factor "produces" its marginal product and that each factor "gets" its marginal product under competitive capitalism. These are not normative assertions. They are factual assertions–which are false. A non-human "agent of production" does not "produce" its marginal product, except in an animistic metaphorical sense. And each hired factor does not "get" a property share of the product, except in the metaphorical sense of the distributive shares picture.

What are the facts unadorned by metaphorical property relations or animism? In a capitalist firm, the facts are that Labor produces the whole product and that Capital gets it. Recognition of those facts does not conflict with the non-ideological analytical use of marginal productivity concepts in price theory.

Let "Labor" be the legal party consisting of all those who work together in a given productive enterprise (regardless of their legal role as employees or working employers). Then Labor is de facto jointly responsible for using up the inputs and for producing the outputs. That is, Labor is de facto responsible for producing the negative product and the positive product, i.e., for producing the whole product. The whole product represents the positive and negative fruits of the labor of the people working in the productive enterprise.

From the legalistic viewpoint, it is the "firm" as a legal party which bears the legal liability for the used-up inputs and which appropriates the produced outputs. Hence the juridical principle (i.e., the labor theory of property) implies that Labor should be the firm, i.e., that the firm should be a self-employment firm or a democratic worker-owned firm [see Ellerman 1990].

These arguments can now, for purposes of comparison, be stated in the simple corn and labor model. Harvest corn and seed corn should be treated as economically distinct commodities. Hence we must use lists or vectors with at least two components:

(Harvest Corn, Seed Corn).

In the productive enterprise, Labor performed the intentional human actions represented as the labor L = aX. These actions used up the inputs AX and produced the outputs X of the firm. Hence the assets and liabilities produced by Labor are:

Labor's Product = (X, –AX).

The labor L = aX is the human activity of producing (X, –AX). But neoclassical economics performs the major conceptual transformation of depicting the human activity of producing (X, –AX) as another "input" in the production process. For the purposes of comparison, that conceptualization of labor is adopted here. Our vectors must be expanded to three components to allow for the labor component:

(Harvest Corn, Seed Corn, Labor Services).

In addition to producing X by using up AX, the workers are also construed as producing and using up the labor services L. This yields the three dimensional version of

Labor's Product = (X, –AX, –L) + (0, 0, L) = (X, –AX, 0).

Since the –L and +L cancel out (when the vectors are added component-wise), the net result for Labor's product is the same as before.

By construing the human activity of production as an input used up in production, we arrive at the vector version of the

Whole Product = (X, –AX, –L).

= (X,0,0) + (0,–AX,–L)

= Positive Product + Negative Product.

The analysis of capitalist production can now be concisely stated. We have seen that Labor is in fact responsible for producing (what has been called)

Labor's Product = (X, –AX, –L) + (0, 0, L)

= Whole Product + Labor Commodity.

Yet Labor only receives title to the labor commodity L which was sold in return for the wages. Labor also produced the whole product, but it was legally appropriated by the employer. That is, the employer sustained the costs for the used-up inputs AX and L, and the employer acquired title to the produced outputs X, so the employer appropriated both the negative and positive product, i.e., the whole product.

Labor produced the whole product, but the employer appropriated it. Thus capitalist production represents "an institutional robbery - a legally established violation of the principle on which property is supposed to rest" [J.B. Clark 1899, 8-9]. How can this happen? Do the legal authorities claim that workers are instruments devoid of responsible agency?

A legal trial (e.g., a property damage suit) can be viewed as an institutional attempt to apply the labor theory of property in its form as the juridical principle of imputation. The trial attempts to ascertain the de facto responsible party so that party can be assigned the de jure responsibility. When no illegality is involved, the legal authorities do not intervene so they make no judgment at all about the de facto responsibility of the workers in normal production. There is a market mechanism of appropriation which takes over when the Law does not intervene. It is a laissez-faire mechanism: let the costs lay where they fall, and then let the party who has borne the costs claim any appropriable outputs.

The employer had purchased (or already owned) the seed corn AX and the labor L. They were not resold, so when those inputs were used up in production, the employer "swallowed" those costs. That is the laissez-faire appropriation of the negative product. Then the same party, the employer, has the legally defensible claim on the outputs X. In that manner, the employer laissez-faire appropriated the whole product (X, –AX, –L).

Three theories concerning the role of Labor under capitalist production have been sketched: marginal productivity theory, the Marxian labor theory of value and exploitation, and the labor theory of property. The principal conclusions are summarized in the following table.

It should be particularly noted that each theory agrees that labor should get what labor produces. The theories each have a different conception of what labor in fact produces and what labor in fact receives. These are factual questions.

Our purpose has been to compare three theories about the role of Labor in capitalist production. Most political economic debate during the last century has been between neo-classical value theory (represented by MP theory) and Marxian value theory. This clash of value theories has not reached the fundamental issues which have to do with property rights, not prices. Hence the two theories have been analyzed from the viewpoint of the labor theory of property.

In its precise modern form, Marxian value theory has emerged as a not particularly insightful "just-price" theory expressing a Marxian version of the old Aristotle-Aquinas interest grumble. Even if one takes it seriously as a just-price theory, it is not a critique of the institution of wage labor but only a critique of unjust wage rates, i.e., the "exploitative" wage rates corresponding to positive rates of profit.

Marginal productivity theory has a proper analytical use. But it is also used as an engine of capitalist apologetics to show that each factor "gets what it produces" in competitive capitalism. But we found this view to be based on two metaphors, the distributive shares picture and the pathetic fallacy. It completely misrepresents the structure of property rights in production as well as the responsible agency (or lack of it) of the various factors of production.

The labor theory of property is a very old theory. But it is also new to political economic debate, a debate which has focused on value theoretic issues for well over a century. Our purpose has been only to present this "old theory" in simple, precise, and modern terms so that it may be compared and contrasted with the Marxian labor theory of value and the neo-classical marginal productivity theory.